Channel-Based Multi-Dimensional Set Partitioning for MIMO Reduced-State Sequence Estimation

A multi-dimensional set partitioning (MDSP) is proposed for reduced-state sequence estimation of multi-path fading multiple input multiple output (MIMO) channels. The proposed MDSP takes channel state information into account when making a set partition because MIMO channels change the distance between symbol vectors in a multi-dimensional constellation. To reduce complexity of the proposed MDSP, we propose an efficient MDSP algorithm which exploits only typical error vectors for partitioning.